Long-Term Antibody Response and Vaccination Efficacy in Patients with COVID-19: A Single Center One-Year Prospective Study from the Czech Republic

Abstract

Background: The diagnosis of SARS-CoV-2 is almost exclusively performed by PCR or antigen detection. The detection of specific antibodies has not yet been considered in official diagnostic guidelines as major laboratory evidence for a case definition. The aim the present study is to analyze antibody responses in outpatient and inpatient cohorts of COVID-19 patients in the Czech Republic over a 12-month period, and assess the potential of antibodies as a diagnostic tool.

Methods: A total of 644 patients was enrolled in the prospective study. IgA, IgM and IgG antibody levels, as well as virus neutralization titers, were analyzed over a 12-month period.

Results: Our study showed low antibody positivity levels at the admission. However, at 2 weeks after infection, 98.75% and 95.00% of hospitalized patients were IgA and IgG positive, respectively. Even in the outpatient cohort characterized by milder disease courses, the IgG antibody response was still sustained at 9 and 12 months. The data show a high correlation between the IgG levels and virus neutralization titers (VNTs). Samples from later time-points showed positive antibody responses after vaccination in both cohorts characterized by high IgG levels and VNT over 1:640. The samples from unvaccinated persons indicated a relatively high level of reinfection at 6.87%.

Conclusions: Our results show that the detection of antibodies against the SARS-CoV-2 shows an increasing sensitivity from week 2 after infection and remains highly positive over the 12-month period. The levels of IgG antibodies correlate significantly with the VNTs. This suggests that the serological data may be a valuable tool in the diagnosis of SARS-CoV-2 infection.

Keywords: ELISA; SARS-CoV-2; dynamics; reinfection; serology; vaccination; virus neutralization test.

Figure 1

The detected antibody levels in the sera of cohort A over the 12-month period expressed as index values. Data are presented as the median ± 95% confidence intervals. The dotted line indicates the time when some patients started to be vaccinated. The nonparametric ANOVA Kruskal–Wallis post hoc Dunn’s multiple comparisons test was used for statistical evaluation. **** represent statistical significance p < 0.0001. The absence of * marks means no significance.

Figure 2

The detected antibody levels in the sera of cohort B over the 12-month period expressed as index values. Data are presented as the median ± 95% confidence intervals. The dotted line indicates the time when some patients started to be vaccinated. The nonparametric ANOVA Kruskal–Wallis post hoc Dunn’s multiple comparisons test was used for the statistical evaluation. Statistical significance is labeled as follows: **** p < 0.0001, ** p < 0.0075; * p < 0.0424.

Figure 3

The dynamics of IgG antibodies in individual patients and the effect of the vaccination and reinfection in patient cohorts A (left column) and B (right column). The charts depict significant increases in antibody levels in groups of vaccinated and non-vaccinated reinfected patients at month 12.

Figure 4

The dynamics of virus neutralization titers in time. The graph illustrates the higher titers in the patients from cohort A in general, and increases in the titers between months 9 and 12 in both cohorts due to vaccination. The statistical evaluation was performed using the nonparametric Mann–Whitney test with the significance set at p < 0.05. **** denotes the statistical significance at p < 0.0001, *** at p = 0.0002 and ** at p = 0.0036.

Figure 5

Correlation of the antibody levels and virus neutralization test. The Spearman’s rank correlation coefficient test shows a positive correlation of specific IgG antibody levels and VNT.

Figure 6

Neutralization of the SARS-CoV-2 Alpha variant with serum samples from patients infected with the original virus variant. The left panel shows a correlation of the virus neutralization titers against the original variant and the Alpha variant by the individual sera. The right panel shows the cumulative values of this analysis. In both panels, decreases in the VNT levels against the Alpha variant are observed. The nonparametric Mann–Whitney test was used for the analysis and **** represents the significance at p < 0.0001.